PREVIOUS studies in this laboratory have shown that cobalt, nickel and cadmium, when injected as the powdered metal into the thigh muscle of rats of the hooded strain, produce a high incidence of rhabdomyosarcomata, whereas a number of other metals, including iron, copper, zinc, manganese, beryllium and tungsten are not carcinogenic under these conditions (Heath, 1956(Heath, , 1960 Heath and Daniel, 1964a, b;Heath, Daniel, Dingle and Webb, 1962). Tumours, which arise at any time from about 3 months after implantation of the active metals, and which metastasize readily in the primary animal, are transplantable into rats of the same strain. A cobalt-induced tumour, for example, has been maintained by transplantation for 13 years and is now in its 137th passage. Corresponding values for a nickel-induced and cadmium-induced tumour are 31 years and 54th transplant, and 6 years and 75th transplant respectively. All of these transplants preserve the characteristics of the primary rhabdomyosarcomata; in general the cobalt primary tumours are the least and the nickel tumours the most differentiated.Since a metallic implant disappears from the site of injection, it is probable that the metal in the dissolved state is one of the main factors in the induction of malignancy. Ions of many of the above mentioned metals, whether carcinogenic or not, are inhibitory to tissue metabolism (e.g. Dingle, Heath, Webb and Daniel, 1962) and are toxic to chick and mammalian cells in culture (Heath, 1953;Daniel, 1961Daniel, , 1964; no common specific biochemical activity of Co2+, Ni2+ and Cd2+ has yet been demonstrated.The work that is reported in this paper is designed to answer the following questions. Firstly, whether the tumours contain high local concentrations of the corresponding metallic ions; secondly, whether these ions are bound firmly and are retained after the dissolution of the metal implants, and, thirdly, whether the carcinogenic cations have a common pattern of distribution, and are associated primarily with a particular intracellular component or organelle.
MATERIAL AND METHODSBiological methods.-Primary tumours were induced by the injection of a suspension of the finely powdered metal (28 mg.; Johnson, Matthey & Co. Ltd., Hatton Garden, London, E.C.1.) in horse serum (0.4 ml.) into the thigh muscle of 2-3 month old rats of the hooded strain as described by Heath (1956) and Heath and Daniel (1964a, b). Tumours for biochemical work were taken usually between 4 and 5 months after the initial implantation. At this stage of development they were free from any significant central necrosis.
ANALYSIS OF METAL-INDUCED TUMOURSTransplants were maintained by serial passage of tissue fragments (about 1.5 mm.3) from the edge of an actively growing tumour into fresh rats at intervals of about 3-4 weeks. Insertion of the tumour tissue into the thigh muscle was made with a large-bore hypodermic needle.All experimental animals were fed ad libitum on Diet No. 86 (North-Eastern Agricultural Co-operative Society Ltd.) rat cubes, and had...